Miscellanea. Comparison of case-control to full cohort analyses under model misspecification
杂论：模型误设下病例对照分析与完整队列分析的比较
We investigate the question of whether or not estimates of relative risk from matched case-control studies well represent the population from which they were drawn. We derive the formula for the asymptotic maximum partial likelihood estimators from nested case-control studies when the Cox proportional hazards model is misspecified and compare these to the corresponding full cohort values under various types of model misspecification, such as covariate omission, mismodelling…